On the fully nonlinear Cauchy problem with small data. II

From MaRDI portal





[For the entire collection see Zbl 0758.00007.] [For part I, cf. Bol. Soc. Bras. Mat., Nova Ser. 20, No. 1, 1-27 (1989; Zbl 0768.35014).] The author studies the Cauchy problem (C.P.) for the equation \((*)\) \(\square u=G(u,u',u)\) with small Cauchy data \((**)\) \(u=\varepsilon u_ 0\), \(\partial_ t u=\varepsilon u\), for \(t=0\), where \(\square\) is the wave operator in \(\mathbb{R}^{1+n}\), \(t=x_ 0\), \(x=(x_ 1,\ldots,x_ n)\). Here \(G\) is \(C^ \infty\) function, vanishing of the order \(2^ d\) at the origin, and the initial data are supposed to be compactly supported. The fact that \(G\) depends on \(u\) causes great difficulties especially for small \(n\). For \(n=3\) and \(n=4\) one has to use a modified energy method [\textit{T. Li} and \textit{Y. Chen}, Commun. Partial Differ. Equations 13, No. 4, 383-422 (1988; Zbl 0662.35071)], estimating \(\bigl\| | D_ x |^{-\sigma}u'(t,x) \bigr \|_{L^ 2}\) where \(| D_ x |^{-\sigma}\) is the potential operator consisting of multiplication of the Fourier transform (with respect to \(x)\) by the function \(\xi \to | \xi |^{-\sigma}\) and \(\sigma \in \bigl[ {1\over 2},1\bigr]\). The main results obtained are: 1) if \(G(u,0,0)=O(u^ 3)\) then the C.P. \((*)-(**)\) has a global \(C^ \infty\) solution for small \(\varepsilon\) if \(n=4\); when \(n=3\) it has a \(C^ \infty\) solution for \(t<\exp(c/ \varepsilon)\), with \(c>0\). 2) Assume \(G\) is \(C^ \infty\), vanishing of order \(2^ d\) at 0; then \((*)-(**)\) has a global \(C^ \infty\) solution for small \(\varepsilon\) if \(n>4\) and for \(n=4\) it has a \(C^ \infty\) solution for \(t<\exp(c/ \varepsilon)\) \((c>0)\). The author explains very carefully the difficulties of the problem, the point where previous methods [cf. \textit{S. Klainerman}, Lect. Appl. Math. 23, Pt. 1, 293-326 (1986; Zbl 0599.35105)] fail. The second paragrah is devoted to the technical results to be used later, in particular a version of the Calderon and Coifman-Meyer commutator estimates. The paper ends with a short paragraph containing a brief discussion about an asymptotically approximating equation, hopefully a first step towards an ``almost global existence theorem for the case \(n=3\).




Cited in
(28)








This page was built for publication: On the fully nonlinear Cauchy problem with small data. II

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4694013)